Temperature-mortality relationship in four subtropical Chinese cities: a time-series study using a distributed lag...

Numerous studies have reported the association between ambient temperature and mortality. However, few multicity studies have been conducted in subtropical regions in developing countries. The present study assessed the health effects of temperature on mortality in four subtropical cities of China.

We used "double threshold-natural cubic spline" distributed lag non-linear model (DLNM) to investigate the cold and hot effects on mortality at different lags in four subtropical cities. Then we conducted a meta-analysis to estimate the overall cold and hot effects on mortality at different lag days.

A U-shaped relationship between temperature and mortality was found in the four cities. Cold effect was delayed and persisted for about 27 days, whereas hot effect was acute and lasted for 3 days. In Changsha, Kunming, Guangzhou and Zhuhai, a 1 °C decrease of temperature under the low threshold was associated with a lag0-27 cumulative relative risk (RR) of 1.061 (95% confidence interval (CI): 1.023-1.099), 1.044 (95% CI: 1.033-1.056), 1.096 (95% CI: 1.075-1.117) and 1.111 (95% CI: 1.078-1.145) for total mortality, respectively. And RR for 1 °C increase of temperature above the hot threshold at the lag0 was 1.020 (95% CI: 1.003-1.037), 1.017 (95% CI: 1.004-1.030), 1.029 (95% CI: 1.020-1.039) and 1.023 (95% CI: 1.004-1.042), respectively. The cold and hot effects were greater among the elderly in Changsha, Guangzhou and Zhuhai. Meta analysis showed that the hot effect decreased gradually with lag days, with the greatest effect at current day (RR=1.023, 95% CI: 1.015-1.031); while the cumulative cold effect increased gradually with lag days, with the highest effect at lag0-27 (RR=1.076, 95% CI: 1.046-1.107).

Both low and high temperatures were associated with increased mortality in the four subtropical Chinese cities, and cold effect was more durable and pronounced than hot effect.